'rotection Technology Series
MOLECULAR SIEVE CONTROL PROCESS IN
SULFURIC ACID PLANTS
Industrial Environmental Research Laboratory
Office of Research and Development
U.S. Efn/ironiiienta! Protection Agency
Research Triangle Park Ncrtfn Carolina 27711
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development,
U.S. Environmental Protection Agency, have been grouped into
five series. These five broad categories were established to
facilitate further development and application of environmental
technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in
related fields. The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL PROTECTION
TECHNOLOGY series. This series describes research performed
to develop and demonstrate instrumentation, equipment and
methodology to repair or prevent environmental degradation from
point and non-point sources of pollution. This work provides the
new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
EPA REVIEW NOTICE
This report has been reviewed by the U.S. Environmental Protection
Agency, and approved for publication. Approval does not signify that
the contents necessarily reflect the views and policies of the Agency, nor
does mention of trade names or commercial products constitute endorse-
ment or recommendation for use.
This document is available to the public through the National
Technical Information Service, Springfield, Virginia 22161.
-------
EPA-600/2-75-066
MOLECULAR SIEVE
CONTROL PROCESS IN
SULFURIC ACID PLANTS
by
D.W. Hissong
Batte lie-Columbus Laboratories
505 King Avenue
Columbus, Ohio 43201
Contract No. 68-02-1323, Task 17
ROAPNo. 21ADH-005
Program Element No. 1AB013
EPA Task Officer: E.J. Wooldridge
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
October 1975
-------
ABSTRACT
An engineering analysis of the applicability of molecular sieve
technology to the control of sulfur dioxide emissions from sulfuric acid
plants has been conducted. Field test data from a commercial plant using
this technology show that after the equivalent of 10 months of operation
the plant is still controlling the sulfur dioxide emissions to well within
the existing Federal and state regulations for sulfuric acid plants. This
plant is also meeting the performance guarantee of the process developer
and vendor (Union Carbide Corporation). The average emission after 10
months was 0.9 Ib SO./ton acid, compared with a most stringent existing
regulation of 4.0 Ib SO./ton acid and a performance guarantee equivalent
to about 1.2 Ib SO./ton acid. The concept of a 2-year sieve life with
acceptable SO. control has not been demonstrated, although there is no
reason to believe it cannot be achieved. Thus, at this point this
application of molecular sieve technology appears technically feasible.
The economic feasibility of molecular sieve technology for this
application was assessed by comparing the total capitalized cost (including
investment and operating cost) of this technology with those of the
Wellman-Lord and dual absorption processes. The capitalized costs for the
three processes are fairly close, and individual acid plant characteristics
will affect the economic choice. The molecular sieve technology is more
competitive for the smaller acid plants and for acid plants which already
have facilities for drying the air needed to regenerate the sieve. Although
for many acid plants the dual absorption process will be the least expensive,
this process is limited in its effectiveness such that for most plants it
cannot meet the regulation of 4 Ib SO./ton acid. Thus, considering the
overall cost and effectiveness, the molecular sieve technology appears
economically feasible for some acid plants.
iii
-------
TABLE OF CONTENTS
Page
INTRODUCTION 1
SULFURIC ACID MANUFACTURING PROCESSES 1
Nature of the Industry I
The Contact Process 2
Plant Size and Type Distribution A
Industry Growth Projection A
Emission Sources 6
Emission Regulations 9
TECHNICAL EVALUATION OF MOLECULAR SIEVE CONTROL TECHNOLOGY .... 10
General Description of Molecular Sieves 10
Use of Molecular Sieves for SO Recovery . . 10
Description of PuraSiv S Process 12
The Coulton Sludge Acid Plant 16
Tests by Union Carbide Corporation 21
Tests by York Research Corporation 22
EMISSION CONTROL ALTERNATIVES 2A
Wellman-Lord Process 2A
Dual Absorption Process 28
Other Emission Control Processes 29
ECONOMIC EVALUATION OF MOLECULAR SIEVE CONTROL TECHNOLOGY .... 33
Basis and Procedures 33
Investment and Operating Costs 37
Capitalized Costs A2
-------
TABLE OF CONTENTS
(Continued)
Page
CONCLUSIONS AND RECOMMENDATIONS 51
REFERENCES 53
APPENDIX A
ACCURACY OF S02 MEASUREMENTS A-l
APPENDIX B
DETAILED COST DATA B-l
LIST OF TABLES
Table 1. Distribution of Contact Sulfuric Acid Plants by Size
and Type 5
Table 2. Contact Sulfuric Acid Plant Tail Gas Emissions 8
Table 3. State Emission Regulations for Sulfuric Acid Plants . . 11
Table 4. Design Basis for PuraSiv S Unit Installed at Coulton
Chemical Company 17
Table 5. Average SO^ Emissions - York Research Tests 23
Table 6. Sulfuric Acid Mist Emissions - York Research Tests ... 25
Table 7. Other Emissions - York Research Tests 26
Table 8. "Feasible" Emission Control Processes for Sulfuric Acid
Plant Tail Gas 32
Table 9. Unit Costs Used in Operating and Maintenance Costs ... 35
Table 10. Costs of PuraSiv S Process 38
Table 11. Costs of Wellman-Lord Process . . 40
Table 12. Costs of Dual Absorption Process for New Acid Plants . . 43
Table 13. Costs of Add-on Dual Absorption Process for Existing
Acid Plants 44
Table 14. Investment and Operating Costs for Existing Acid
Plants 45
vi
-------
LIST OF TABLES
(Continued)
Page
Table 15. Equations for Investment and Operating Costs 46
Table 16. Capitalized Costs for Existing Acid Plants 47
LIST OF FIGURES
Figure 1. Flow Sheet of Sulfur-Burning Contact Sulfuric Acid
Plant 3
Figure 2. Sulfur Dioxide Emissions at Various Conversion
Efficiencies 7
Figure 3. Relationship of Conversion Efficiency to SO.- Concen-
tration in Exit Gases 7
Figure 4. Flow Sheet of PuraSiv S Process with Acid-Dried Air
Regeneration 13
Figure 5. Flow Sheet of PuraSiv S Process with Sieve-Dried Air
Regeneration 15
Figure 6. S02 Adsorber 18
Figure 7. Air Dryer 19
Figure 8. Brink's Mist Eliminator 20
Figure 9. Flow Sheet of Wellman-Lord Process 27
Figure 10. Flow Sheet of Dual Absorption System for New Acid
Plants 30
Figure 11. Flow Sheet of Add-on Dual Absorption Process Using
External Heat Source 31
Figure 12. Capitalized Costs for SO Control of Dry Gas Acid
Plants 48
Figure 13. Capitalized Costs for SO- Control of Wet Gas Acid
Plants 49
vii
-------
ENGINEERING ANALYSIS OF A MOLECULAR SIEVE
CONTROL PROCESS IN SULFURIC ACID PLANTS
by
D. W. Hissong
INTRODUCTION
The manufacture of sulfuric acid results in emissions of sulfur
oxides and sulfuric acid mist. The emissions of acid mist can be reduced
to meet the anticipated regulations through the use of efficient mist
eliminators. A number of technologies are being considered for controlling
the sulfur oxide emissions. Among the alternatives is a process involving
adsorption of the sulfur oxides onto a molecular sieve followed by cyclic
regeneration of the adsorbent and recovery of the sulfur oxides. This
process is called the PuraSiv S process and has been developed by Union
Carbide Corporation. Union Carbide claims that this process can reduce
the average concentration of sulfur oxides in the effluent gas stream to
less than 100 ppm by volume, which corresponds to an emission rate well
below the present or anticipated regulations.
In an effort to determine the technical and economic feasibility
of molecular sieve control systems for this application, the U.S. Environ-
mental Protection Agency sponsored this study. The objectives were to
determine
• The technical feasibility of this control system by
analyzing data independently obtained from an
operating unit as well as information obtained from
the system developer
• The economic feasibility of the system by comparing
it with other commercially available control
processes.
SULFURIC ACID MANUFACTURING PROCESSES
Nature of the Industry
Sulfuric acid is a very basic industrial chemical, being involved
in countless manufacturing operations. It is of such paramount significance
-------
that the consumption of this chemical has been used as a business indicator
and a measure of technical development of nations. The production of
sulfuric acid in the United States in 1973 was 28.8 million metric tons.
About half of this acid was used in the manufacture of phosphate fertilizers.
Essentially all the sulfuric acid in the United States is now
produced by the Contact Process, less than 0.4 percent being produced by
the older Chamber Process. There are currently 152 plants in this country
using the Contact Process and only five plants using the Chamber Process.
The Contact Process
(2)
A flow sheet for the Contact Process is shown in Figure 1.
The source of sulfur for the process can be elemental sulfur, spent sulfuric
acid (sludge), hydrogen sulfide, or other sulfur compounds. This source is
burned with dry air in a furnace to produce sulfur dioxide (S0«). The
latter is further oxidized to sulfur trioxide (SO.) over a catalyst which
usually consists of supported vanadium pentoxide. The resulting gases
are cooled and then introduced into an absorption tower in which the S0_
is absorbed in water to form sulfuric acid. The reactions involved are
as follows (writing the sulfur source as S).
Burner S + 0 -»• SO
Oxidation reactor SO + 1/2 0_ ->• SO
Absorption tower SO + HO -»• H^SO,
The product is usually marketed as 93-98 percent H^SO, solutions or oleums,
which are solutions of S0_ in H SO .
Contact sulfuric acid plants are usually classified as either
"sulfur burning plants" or "wet gas plants". However, the significant
distinction for this study is between the wet gas plants and what one
might call "dry gas plants". About three-fourths of the sulfuric plants
in this country are of the dry type, which means that the gas leaving the
burner is kept as dry as possible by drying the input streams to the burner.
This drying is done with some of the concentrated acid (98 weight percent)
and is necessary if the plant is to produce a concentrated acid product.
-------
4
SULFUR
FURNACE
BOILER
FEED •
WATER
K irtm i h
t
X
5
STEAM
BO
3
r A OLEUM
,, . L TOWER (C
r i
ILER. BOILER ry.'S// CONVERTER • 1* ~
\(\ — ^ — f\(*\ WITH ^ ^
'SM — j — »VV » . INTERCOOLERS N/
— ^/ xV/ • xV
X N
J i
v,y//x *-- -" !
1
ECONOMIZER x -v I
DRYING
TOWER
COOLER
i)
•r^,^ ' ' i
1 ABSORPT
I TOWER
1'
•
i X
J i ,
JPTIONAL).
i
i
i
i
-Hr-(*\^ ' OU^1
" V_X PRODUCT
.TAIL
ON
1
JL COOLER
$
ACID
„ PRODUCTS
FIGURE 1. FLOW SHEET OF SULFUR-BURNING CONTACT SULFURIC ACID PLANT
-------
All plants which feed exclusively elemental sulfur or sulfur ores are
operated dry. However, plants feeding a hydrogen sulfide stream can be
operated either wet or dry. There are two consequences of wet operation
which are important here
• The volume of the tail gases to be treated is greater
for wet gas plants than for dry gas plants
• Wet gas plants do not have available a source of dried
air for regenerating the molecular sieve.
Plant Size and Type Distribution
Of the 152 contact sulfuric acid plants in the United States, 112
are dry gas plants and 40 are wet gas plants. The total capacity of the
dry gas plants is 74,310 tons* (67,400 metric tons) per day and that of the
wet gas plants is 26,205 tons (23,770 metric tons) per day. Table 1 shows
the size distribution for both types of plants. ' The plant capacities
range from 15 to 2000 tons per day, with the average capacity being about
660 tons per day.
The 1973 production rate of 31.8 million tons (28.8 million metric
tons) per year corresponds to about 87 percent of the capacity listed in
Table 1.
Industry Growth Projection
The annual production of sulfuric acid in the United States is
projected to increase to about 31 million metric tons in 1976 and 37 million
metrj
that
metric tons in 1980. These projections are based upon the assumptions
e The production of acid from smelter gases remains
constant at the present level of 3.4 million metric
tons per year
« The fertilizer uses of sulfuric acid increase at a
rate of 6 percent per year and the nonfertilizer
uses at 4 percent per year.
* "Tons" refers to the short tons (2000 Ib) normally used in this industry.
A "metric ton" is 2,205 Ib.
-------
TABLE 1. DISTRIBUTION OF CONTACT,SULFUR!C ACID
PLANTS BY SIZE AND TYPEU'
Tons per
Size
Range
15- 165
170- 225
245- 285
300- 430
450- 570
600- 615
645- 750
- 800
- 830
900- 950
1000-1040
-1200
1300-1450
1460-1550
-1600
1750-1900
-2000
Day
Average
Size for
Range
106
200
260
360
500
600
700
800
830
900
1015
1200
1400
1500
1600
1850
2000
Number of Plants
of Each Type
Dry Gas
15
13
6
19
14
4
7
—
1
9
2
1
5
4
6
4
2
112
Wet Gas
5
2
5
11
4
2
4
2
—
—
1
—
1
—
1
—
_2
40
Percent of
Capacity,,.
of Category v '
Dry Gas Wet Gas
2.1
3.5
2.1
8.9
10.0
4.9
6.4
—
1.1
10.9
2.7
1.6
9.4
8.1
12.9
10.0
5.4
2.5
2.4
4.9
15.8
7.6
4.6
10.7
6.1
—
7.1
3.8
—
5.5
—
6.1
—
22.9
100.0 100.0
(a) Plants existing in 1974 that went on-stream prior to 1973.
(b) Total capacity in dry process plants is 74,310 tons per day;
for wet process plants it is 26,205 tons per day.
-------
6
Emission Sources
The major source of emissions from contact sulfuric acid plants
is the exit gas, or tail gas, from the absorber.* This gas contains
unreacted S02> unabsorbed S03> and sulfuric acid,mist. In the atmosphere
the S03 will be converted into sulfuric acid mist by hydrolyzing with
moisture in the air. The most abundant pollutant in the tail gas is SO-.
The amount of SCL in the tail gas depends upon the conversion
efficiency of the oxidation reactors, which in turn depends on a number of
factors including
9 Concentration of sulfur dioxide in the gases entering
the converter and the ratio of oxygen to sulfur dioxide
particularly in the last converter stage
9 Number of converter stages
• Volume and distribution of catalyst in various converter
stages
• Catalyst efficiency (depends on catalyst type and age)
0 Uniformity of gas composition
e Impurities in the entering gas
• Temperature control at various points in the converter
(this depends in part on having properly sized inter-
stage gas cooling equipment).
Figure 2 shows the relationship of the conversion efficiency to the SO.
emission in pounds per ton of acid** produced. Most contact sulfur
acid plants are purchased with guarantees of 96 to 98 percent conversion
efficiency. This corresponds to emissions of about 25 to 55 pounds of S0_
per ton of acid. For comparison, the New Source Performance Standard is
4 pounds of SO- per ton of acid.
Figure 3 shows the relationship of the conversion efficiency to
the S0» concentration in the tail gas based on 26 commercial
(4)
acid plants. The average for these plants was about 97 percent
conversion efficiency and an S02 concentration of about 2700 ppm in the
tail gas. Table 2 summarizes the data on which this Figure was based.
Table 2 also summarizes the concentrations and emission rates
of sulfuric acid mist for the plants. These values vary widely depending
on details of the design and operating conditions. Many plants which do
* The other sources of atmospheric emissions include gases vented from
storage tanks, tank cars, and tank trucks during filling operations and
from acid concentrators. Although few data are available on emissions
from these sources, they are evidently at least an order of magnitude
less than the tail gas emissions.
** kg SO /metric ton acid = (0.5)(lb SO /ton acid).
-------
93 94 95 96
Conversion of SO,
97
to SO,
98
99
100
/. WV WN/.f »
FIGURE 2. SULFUR DIOXIDE EMISSIONS AT VARIOUS CONVERSION EFFICIENCIES
(Per Ton of Equivalent 100% SO Produced)
Conversion of S0_ to SO., %
a s s . s s
ooooo
\
AVCPAGE
\
\
•_
> 0.1 0.2 0.1 0.4 OJ 0.
SO- in Exit Gas, %
FIGURE 3. RELATIONSHIP OF CONVERSION EFFICIENCY TO
S02 CONCENTRATION IN EXIT GASES
-------
TABLE 2. CONTACT SULFUKIC ACID PLANT TAIL GAS EMISSIONS
Type of Plant
Sulfur burning
Air dilution
No air dilution
Air dilution
No air dilution
Average
Combination
Spent Acid H.,8
X
X
X
X X
X X
X X
X X
X
X
Tail Gas Rate
Controls (103 scf/T)
-1 None 88. 2
-2
-3
-4
-5
-6
106.3
91.2
85.6
83.1
93.6
-1 None 73.8
-2
-3
-4
-5
73.9
82.3
86.4
79.0
-6 T 98.9
Mist Elim. 86.9
Mist Elim. 71.0
85.7
S
X -1 None 129.6
X -2 1 93.9
-3 » 99.2
X -1 Mist Elim. 78.9
X -2
X -3
X -4
X -5
X -6
75.8
81.6
88.4
102.2
" 158.4
SO- Cone.
(volume percent)
0.20
0.31
0.23
0.25
0.28
0.20
0.53
0.19
0.24
0.40
0.25
0.42
0.14
0.19
0.27
0.37
0.20
0.34
0.32
0.17
0.32
0.18
0.20
SO. Emission
(lb/T) •
31.3
58.5
36.7
41.5
41.5
34.0
70.2
24.3
35.2
61.6
34.8
56.6
21.6
24.7
40.6
85.2
33.1
60.0
36.5
22.6
46.7
27.6
36.8
Mist Cone.
(mg/scf)
2.4
19.6
3.0
9.2
9.2
3.0
3.7
2.3
9.5
6.1
1.9
37.3
6.5
13.5
8.4
11.0
10.2
0.3
0.3
0.5-2.0
26.3
Mist Emission
(lb/T)
0.46
4.6
0.57
1.7
0.80
0.58
0.35
1.7
1.2
0.32
6.3
2.4
2.3
2.2
0.03
0.05
0.07-0.3
0.22
00
Note: scf
cubic feet at 32 F and 1 atm; T = short ton of 100 percent H.SO,.
-------
9
not have an external mist eliminator do have some mist elimination
facilities built into the top of the absorber. Also, impingement in the
stack and lines leading to it may reduce the mist content of the gas
before it is exhausted to the atmosphere. The mist emission rates for the
sulfur burning plants listed in Table 2 vary from 0.3 to 6.3 pounds per
ton of acid produced. For comparison, the New Source Performance Standard
is 0.15 pounds per ton of acid.
Because the sampling and analytical techniques for SO. are more
complex than for SO. and acid mist, the test data on SO. are more limited.
For the plants in Table 2 for which SO, concentrations were reported the
values ranged from 0.5 to 48 milligrams per cubic foot. This S0_ ultimately
adds to the total acid mist emission of the plant. The range of concen-
trations just cited corresponds to contributions of 7 to 85 percent of
the total acid mist emissions from the plants involved. Thus, the unabsorbed
SO., can comprise an appreciable part of the total acid mist emission.
In the absence of further controls, the emission of SO. from
sulfuric acid plants, excluding smelter gas plants, is estimated to be
about 630,000 metric tons in 1976 and 764,000 metric tons in 1980. The
corresponding emissions of acid mist are 19,000 metric tons in 1976 and
(3)
24,000 metric tons in 1980. *" '
The volume of the tail gas stream is important in designing
emission control equipment. One can see from Table 2 that the average tail
gas rate for sulfur burning plants was 85,700 scf per ton of acid produced.
This value was used for the dry gas plants. For wet gas plants the tail
(3)
gas rate per ton of acid is about 1.5 times that for dry gas plants.
For wet gas plants the SO. concentration in the tail gas is about
/ 3\
3500 ppm compared to the previously quoted 2700 ppm for dry gas plants. '
This higher concentration plus the higher flow rate means that the S02 emission
per unit of acid produced is about twice that for dry gas plants.
Emission Regulations
As mentioned previously, the New Source Performance Standards
for sulfuric acid plants limit the emissions to 4 pounds of SO- and 0.15
pounds of acid mist per ton of acid produced (2 kg of S02 and 0.075 kg of
acid mist per metric ton of acid). These limits are based on 2-hour
-------
10
average conditions and apply to plants where the construction or modification
started after August 17, 1971. They do not apply to smelter gas plants and
other situations in which the acid is made primarily as a means of controlling
SO pollution.
For sulfuric acid plants built before August 17, 1971, the state
regulations apply. Table 3 lists the regulations proposed in the State
Implementation Plans. One can see that most states limit the SO- emissions
to 6.5 pounds per ton of acid and the acid mist emissions to 0.5 pounds per
ton of acid.
TECHNICAL EVALUATION OF MOLECULAR SIEVE CONTROL TECHNOLOGY
General Description of Molecular Sieves
The term "molecular sieve" was originated by J. W. McBain to
describe fine porous solid materials which have a structure of channels
and voids which permits them to act as sieves on a molecular scale. A
molecular sieve has pores of a very uniform size which are uniquely
determined by the unit structure of the crystal. These pores will completely
exclude molecules which are larger than their diameter.
An important type of. molecular sieve is the zeolites, which are
crystalline, hydrated aluminosilicates of the alkali, and alkaline earth
elements. The structure of zeolites consists of a three-dimensional
network of SiO, and A10, tetrahedra linked together by sharing all the
oxygen atoms. There are 34 known species of natural zeolites and about
100 types of synthetic zeolites, although only a few have practical
significance at the present time.
The selectivity of a molecular sieve for a particular adsorbate
depends not only on the dimensions of the interstitial spaces in the
sieve crystal, but also on the polarity of the adsorbate, and, in the case
of hydrocarbons, the extent of carbon-bond saturation of the adsorbate.
Use of Molecular Sieves for SO,, Recovery
In 1963 the Bureau of Mines reported on a study on the use of
molecular sieves for recovering S07 from industrial gases. The impetus
for this study included both air pollution abatement and supplementing
production of sulfur from "secondary" mineral sources such as gypsum. A number of
-------
11
TABLE 3. STATE EMISSION REGULATIONS FOR SULFURIC ACID PLANTS
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Maximum
Ib/ton 100
so2
6.5
6.5
6.5
6.5
6.5
6.5
6.5
14.2
6.5
4.0
. 10.0
6.5
6.5
40.0
6.5
6.5
30.0
6.5
6.5
6.5
6.5
27.0
6.5
6.5
6.5
6.5
6.5
6.5
6.5
6.5
6.5
40.0
6.5
27.0
6.5
6.5
4.0
6.5
6.5
6.5
4.0
6.5
6.5
30.0
8.0
6.5
27.0
6.5
30.0
4.0
4.0
Emission,
percent HrtSO,
Acid MisE
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.15
0.15
0.5
0.5
0.15
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.7
1.7
0.5
0.5
0.5
0.5
0.5
0.18
1.88
0.5
0.5
0.5
0.5
0.5
0.15
0.5
0.5
0.5
0.5
0.5
0.5
0.5
1.6
0.5
0.9
0.5
0.5
0.5
0.15
-------
12
synthetic zeolites were tested as selective adsorbents for recovering S0_
from various synthetic gases and from gases produced by decomposing
gypsum. Three zeolites were found which had capacities considered adequate
for this application. These capacities were up to 29 grams of SO per
100 grams of adsorbent. Regeneration studies showed that over 90 percent
of the SO- desorbed from the loaded beds was of sufficient purity for
direct liquefaction. However, all the zeolites tested lost adsorptive
capacity with repeated use.
The Union Carbide Corporation has been working on the use of
molecular sieves for SO removal for over 5 years. This work has included
laboratory determinations of SO selectivities and equilibrium loadings,
mass transfer rate measurements, and cyclic life tests. This work led to
a pilot scale test program and thereby to the development of the PuraSiv
S Process to be considered here.
Description of PuraSiv S Process*
The PuraSiv S Process, developed by Union Carbide Corporation,
is a fixed bed adsorption system for concentrating and removing S0? from
sulfuric acid plant tail gas. The adsorbent used is a synthetic molecular
sieve which Union Carbide has developed for this purpose. Upon regeneration
of the molecular sieve, the S0_ is recovered in concentrated form and is
recycled to the acid plant. Theoretically, the recovered S02 should contribute
a 2-3 percent increase in acid production rate, although this has not yet been
verified in the commercial test unit. The process also removes and recovers
SO., and acid mist from the tail gas.
Regeneration of the loaded adsorbent bed is accomplished by a
thermal swing, using hot air to flush out the S0« as a concentrated stream.
There are two versions of the process which differ in the source of the
air for regeneration. If a source of acid-dried air is available, such air
will be used for the regeneration and the process flow sheet will be as
in Figure 4. If no such source is available, the regeneration gas will be
supplied by drying ambient air with an additional molecular sieve. The flow
* Based on information obtained from Union Carbide Corporation and
Reference 8.
-------
4,451
(20,433)
Acid-dried
air
Mist
eliminator
4,451
(20,435) Blower
Cooler
17,996
(81,170)
Mist
eliminator
18,000
(81,225)
Blower
Cooler
Adsorbers
Adsorption
Note:
Switching valves for adsorbers
are not shown.
Numbers are ft /min at 60 F
and 1 atm (Ib/hr) for typical
design. Inlet
2200 ppm.
SO cone
S02 rich
gas
Cooling
Vo£ 4,490
Cooler(20,823)
t_
A
Heater
Regeneration
(Heating)
TT
Fuel Air
Treated
tail gas
17,956
: (80,780)
FIGURE 4. FLOW SHEET OF PURASIV S PROCESS WITH ACID-DRIED AIR REGENERATION
-------
14
sheet for this version is shown in Figure 5. Normally, a source of acid-
dried air will be available in "dry gas" sulfuric acid plants but not in
"wet gas" plants.
In both versions the tail gas passes through a blower, a-cooler,
and a high efficiency mist eliminator which decreases the mist concentration
to less than 0.1 mg/scf. The mist eliminator operates in such a way that
one stage of H_SO, vapor absorption by the acid wetted elements takes
place. The tail gas then passes through a molecular sieve bed for selective
adsorption of SO,-. The molecular sieve provides the most effective perfor-
mance and longest life when the tail gas is bone dry. This is accomplished
by drying the gas with a dessicant grade molecular sieve located in the
same adsorber vessel in a compound bed arrangement. The treated tail gas
is discharged to the atmosphere.
In the acid-dried regeneration version (Figure 4), air from
the acid plant drying tower, which usually contains 50-100 ppm of water,
passes through a blower, a cooler, and a high efficiency mist eliminator.
This cooled regeneration gas is first used to cool the molecular sieve bed
which is hot from a previous regeneration step. This heats the regeneration
gas and reduces its water content to <1 ppmv. This regeneration gas then
passes through a furnace where make-up heat is added and then is used to
regenerate the loaded molecular sieve bed. The hot gas strips the SO. and
water from the adsorbent. The resulting S0~-rich gas is cooled and then
recycled to the acid plant.
The four adsorber vessels shown in Figure 4 are provided with
the necessary switching valves for the cyclic operation. At any given
time, two adsorbers are on the adsorption step, one is on the cooling
step, and one is on the regeneration (heating) step.
In the sieve-dried regeneration version (Figure 5), an additional
pair of adsorbers is used to provide the dry air for regeneration. Ambient
air passes through a blower, a cooler, a knock out pot, and then an adsorber
containing a molecular sieve dessicant. This reduces the water content of
the air to <1 ppmv. This dry air is then used to regenerate two adsorbers
which are operating on a staggered cycle. Half of the regeneration gas is
heated and passes through one of the adsorbers until a portion of the bed
-------
Ambient X~*N
air /^*\ — (\) — •
( ) X.
6,224 V-^ Cooler
(28,202)
J^rHSh
^
f
6,116
(27,892)
**>>
Knock-out
DOt
t Adsorbers
V
18,000 V-K Cooler
(84,527)
Note: Switching valves for
shown.
17,996
(84,472)
i
f
A
ruel
Heater
f
Air
/ V Heater
6,010
(27,589)
Cooler 1
Adsorbers
Mist
eliminator
adsorbers are not
i
1
Adsorpt on
t
V**>\ Fue
r\ *
1
-^V '
4
Air
i
Cooling
f
**~^L
f~^\
W
SO 2 rich
X^ g°s
Cooler
6,055
(28,035)
Regeneration
(Heating)
-
I
Nunbers are ft /min at 60 F and 1 atm
(Ib/hr) for typical design. Inlet SO
«*^&M <*_ocrt/%^^_^ £*
cone = 2500 ppm.
FIGURE 5. FLOW SHEET OF PURASIV S PROCESS WITH SIEVE-DRIED AIR REGENERATION
Treated tail gas
18,057
(84,330)
-------
16
is at the regeneration temperature. This regeneration gas is then replaced
with the other half of the regeneration gas, this being the portion which
was not heated but rather was cooled. This cool gas cools the bed behind
the hot zone and in effect "pushes" the hot zone through the bed. This is
referred to as a "thermal pulse" regeneration. When the hot zone "goes out
of the end" of the adsorber the bed is cool and ready for the next adsorption
step. The effluent streams from the two adsorbers are combined, cooled, and
recycled to the acid plant. This method of regenerating two adsorbers in
parallel on a staggered cycle results in a more efficient utilization of the
regeneration gas heater for the sieve-dried regeneration system.
The adsorber which was used to dry the air is periodically
regenerated by the "thermal pulse" technique. A portion of the treated tail
gas is used for this regeneration, and the desorbed water passes directly
to the stack with the treated tail gas.
The Coulton Sludge Acid Plant
A PuraSiv S unit has been installed on one of two contact
sulfuric acid plants (the "B" plant) at the Oregon, Ohio, plant of Coulton
Chemical Corporation. This plant processes a mixture of alkylation spent
acid (sludge) and refinery H S off gas. The plant is designed for a peak
production rate of 200 tons per day of 99 percent H2SO,. The fuel used is
oil.
The design basis for the Coulton PuraSiv S unit is given in
Table 4. The unit was designed to reduce the S0» concentration of 9,450
scfm of tail gas from 3500 to less than 100 ppm. The unit includes a
Brinks mist eliminator, two adsorbers for SO adsorption/desorption, and
2. ;
two adsorbers for drying the regeneration air. Pictures of the major
equipment items are shown in Figures 6, 7, and 8.
The Coulton unit went on stream February 4, 1973. The availability
of the unit since that time has been very good. Immediately following
startup Union Carbide performed a 6-month emission study on the system.
This study showed an average SO- collection efficiency of 99.15 percent.
Additional data from this study are discussed in the next section.
-------
17
TABLE 4. DESIGN BASIS FOR PURASIV S UNIT INSTALLED
AT COULTON CHEMICAL COMPANY
Normal
Flow
High
Flow
Battery Limits Tail Gas Feed Conditions:
Flow rate, (60 F, 14.7 psia) maximum
Temperature, maximum
Pressure
Composition, maximum concentration
by volume
SO
S°3
NO
H-SO, (vapor + mist)
4
Other minerals acids (HC1, HF, etc.)
Organic sulfides (H2S, CS2> etc.)
Hydrocarbons
Halogens (F,,, C12, etc.)
9,450 scfm
170 F
17.1 psia
3,500 ppm
10 ppm
10 ppm
100 ppm
200 ppm
7.5 percent
10.0 percent
0.2 ppm
0.5 ppm
1.0 ppm
0.5 ppm
Balance
10,446 scfm
170 F
17.5 psia
2,800 ppm
10 ppm
10 ppm
100 ppm
200 ppm
7.5 percent
10.0 percent
0.2 ppm
0.5 ppm
1.0 ppm
0.5 ppm
Balance
-------
18
i;
FIGURE 6. S02 ADSORBER
-------
19
FIGURE 7. AIR DRYER
-------
20
FIGURES. BRINK'S MIST ELIMINATOR
-------
21
In November of 1973 the molecular sieve in the Coulton unit
was replaced with a newer, more durable, second-generation sieve material.
In February of 1975, after this sieve had seen approximately 10 months of
equivalent usage, a test program was conducted by the York Research
Corporation. This study showed an average SCL collection efficiency of
98.15 percent. The results of this test program are further discussed
in a future section.
Tests by Union Carbide Corporation
(8)
The following data have been reported by Union Carbide for
a portion of their test program. About 2 months after startup, an
extensive field evaluation of the Coulton unit was carried out by Union
Carbide over a continuous period of 10 days. During this period, the tail
gas flow rate varied from 8,320 to 9,860 scfm and the inlet SCL concen-
tration from 2500 to 5000 ppm. In spite of these variations the unit
provided an average daily effluent SO- concentration of 15-25 ppm. This is
equivalent to an emission rate of less than 0.4 pounds of SCL per ton of
acid and to an S0_ recovery rate of over 99 percent.
Union Carbide has guaranteed a maximum average effluent concen-
tration of 100 ppm SCL for this unit. Based on the design gas flow rate
(9,450 scfm), this corresponds to a maximum emission rate of about 1.2
pounds of SO. per ton of acid. Thus, the above data indicate that the
Coulton unit met the vendor's performance guarantee during these tests.
-------
22
Tests by York Research Corporation
Sulfur Dioxide
During the 4-week test program conducted by York Research Corpor-
ation, the SO- concentrations in the inlet and outlet ducts of the unit
were monitored with a Du Pont photometric analyzer. A total of 118 complete
cycles of sieve operation were recorded during which no malfunctions or
upsets occurred.
In order to determine the accuracy of the Du Pont analyzer,
exit gas samples were periodically extracted and analyzed on-site using EPA
Method 6. The results of these comparisons, which are discussed in the
Appendix, showed that the accuracy of the Du Pont analyzer was within the limits
specified by the EPA for S0« emissions testing.
The complete results of this test program are given in the report
(9)
prepared by York Research Corporation. These results will only be
summarized and analyzed here.
Table 5 summarizes the results of the SO emissions tests. The
results are given separately for the two adsorber vessels and for three
inlet SO. concentration ranges. With respect to the latter, "design
conditions" refers to the range of 4000-5000 ppm S0_.
The results show that adsorber Al was operating at a somewhat
lower efficiency than adsorber A2, this being due to a minor difference in
design. The average emissions were 0.97 Ib S0_/ton of acid for Al and 0.78
Ib S09/ton of acid for A2. Both values are considerably less than the New
Source Performance Standard of 4 Ib S0?/ton of acid.
The results also show that the emissions increase with increasing
inlet S00 concentration. For adsorber A2, the emission was 0.67 Ib S00/
i. £
ton of acid for <4000 ppm SO? inlet, 0.87 Ib/ton for 4000-4500 ppm, and 1.04
Ib/ton for >4500 ppm.
The time span of this study was too short to detect any changes
in the emissions with time. York recommended that a series of studies at
periodic intervals be conducted for this purpose, possibly when a new
molecular sieve system is installed on the "A" plant at Coulton. For a
new unit such as this, York recommended a 2-month monitoring program just
after startup and succeeding 1-inonth programs at 6-month intervals.
-------
23
TABLE 5. AVERAGE S02 EMISSIONS - YORK RESEARCH TESTS
Number of Cycles Measured
At design conditions
At low inlet concentrations
At high inlet concentrations
TOTAL
Average Emission (lb/SO«/T H^SO,)
i. 2. 4""
At design conditions
At low inlet concentrations
At high inlet concentrations
WEIGHTED AVERAGE
Average Emission Rate (Ib S0n/hr)
At design conditions
At low inlet concentrations
At high inlet concentrations
WEIGHTED AVERAGE
Adsorber Al
18
35
6
59
1.13
0.81
1.36
0.97
7.55
5.07
9.09
6.24
Adsorber A2
16
34
9
59
0.87
0.67
1.04
0.78
5.80
4.20
6.92
5.05
Both Adsorbers
34 Total
69 Total
15 Total
118 Total
1.00 Average
0. 74 Average
1.20 Average
0.87 Average
6.68 Average
4.64 Average
8.00 Average
5.64 Average
-------
24
Other Pollutants
In addition to the S0? monitoring program, a series of wet chemical
analyses was performed for sulfuric acid mist, total acid, chloride, sulfide,
hydrocarbons, nitrogen oxides, moisture, carbon dioxide, and oxygen. The
results of the acid mist analyses are summarized in Table 6 and those of
other analyses in Table 7.
The emissions of sulfuric acid mist were less than 0.025 Ib/ton of
acid, which is far less than the New Source Performance Standard of 0.15
Ib/ton of acid.
The outlet concentration was less than the inlet concentration
for acid mist, chloride, hydrocarbons, and nitrogen oxides, indicating that
the molecular sieve adsorbed these species to some extent. No samples of
the regeneration off gas were taken, so the extent to which these species
were desorbed from the sieve is unknown.
EMISSION CONTROL ALTERNATIVES
In the next section, the economic feasibility of molecular sieve
emission control technology for sulfuric acid plants will be assessed by
comparing this technology with other commercially available control alternatives,
The two processes selected for this comparison are the Wellman-Lord Process
and the dual absorption process. In this section these two processes will
be described and other possible emission control alternatives will be briefly
discussed.
Wellman-Lord Process
The- Wellman-Lord Process is a regenerative SO. removal process
is available from Davy Powergas, Inc. A flow sheet for this process
is shown in Figure 9. In this process the tail gas is scrubbed with an
aqueous slurry containing mainly sodium sulfite (Na_SO_) but also some
sodium bisulfite (NaHSO ), sodium sulfate (Na0SO,), sodium hydroxide (NaOH),
and sodium thiosulfate (Na.S 0 ). The absorber contains two to five
-------
TABLE 6. SULFURIC ACID MIST EMISSIONS - YORK RESEARCH TESTS
Time Mist Concentration (mg/scf) -"Mist Emission (Ib/T H,SO,) Mist Emission Rate (Ib/hr) Percent Emission
Date
2/26
2/26
2/27
Average for
Inlet
1117
1315
1032
2/20
Outlet
1120
1315
1030
through 2/27(a)
Inlet
0.167
0.206
0.158
0.215
Outlet
0;.121
' 0.141 '
0.099 -
0.071
Inlet
0.025
••-',. . .. . Jp;032
::b.024
•./•: :.; -0.034
Outlet
0.018
.v' 0.021 '••..;
.0.015
0.011
Inlet
.0.168
0.215
0.159
0.216
Outlet
0.121
0.141
0.099
0.071
Reduction
28
34
38
67
(a) Five inlet measurements and six outlet measurements.
-------
TABLE 7. OTHER EMISSIONS - YORK RESEARCH TESTS
Average Concentration (mg/scf )
Specie
Total acid (as H SO.)
Sulfide (as CS )
Chloride (as Cl~)
Hydrocarbons (as hexane)
NO
X
Moisture
Dates
2/6 - 2/25
2/14 - 2/21
2/6 - 2/25
2/7
2/6 & 2/25
2/12 - 2/21
Inlet
5.44
0.53
0.1128
0.00095
0.385
Outlet
0.97
0.59
0.0015
0.00075
0.089
(a)
Average Emission Rate (Ib/hr)
Inlet
7.0
0.146
4.14
0.90
Outlet
1.3(b)
0.002
2.85
0.71
(a) Each value is the average of 6 to 11 measurements.
(b) Estimated from gas flow rate calculated from chloride data (same sampling times).
-------
Make-up
NaOH
Mist
eliminator
Clean tail gas
Absorber
Toil gas
Make-up
»•
water
Venturi
scrubber
Dissolving
tank
Surge
tank
S02 rich gas
*
N>
Condensate
FIGURE 9. FLOW SHEET OF WELLMAN-LORD PROCESS
-------
28
individually recirculated stages. The main reaction in the absorber is
the conversion of the sulfite into the bisulfite as a result of S0?
absorption:
S02(g) + Na2S03(aq.) +
The scrubbed gas leaves the absorber through a mist eliminator.
The liquid solution from the absorber is preheated with hot
water and then goes to an evaporator, where the heat supplied causes the
conversion of the bisulfite into sulfite, the latter crystallizing out
because of its low solubility. The sulfite slurry from the evaporator
goes to a dissolving tank where makeup caustic soda (NaOH) or soda ash
(Na,,CO ) are added. The resultant slurry is recycled to the absorber.
The S0?-rich vapor stream from the evaporator goes to a primary condenser
which removes 80-90 percent of the water. The resulting condensate is steam
stripped to remove the dissolved SO,.. The overhead vapors from the stripper
are mixed with the vapor from the primary condenser and then pass through a
secondary condenser. Additional water condensed at this point is returned
to the top of the stripper. The vapor stream from the secondary condenser
contains about 85 percent SO- and 15 percent water.
Dual Absorption Process
The dual absorption process is really an alternate design of the
sulfuric acid plant which reduces the S0_ emissions. In this process the
tail gas from the primary absorber is reheated and returned to the converter
where more of the SO is converted to SO . The gas then goes to a secondary
absorber where the SO, is absorbed to form more acid. Usually the tail
gas from the primary absorber only goes through the last stage of the
converter. The additional conversion of SO,, to SO, is possible because the
£• O
tail gas from the primary absorber has a higher 0^/SO. ratio than the gas
originally sent to the converter (typically a ratio of >4 instead of 1.5).
However, dual absorption cannot reduce the S0? concentration in the tail
gas below about 500 ppm.
The specifics of the flow sheet for dual absorption depend on
the application, the key feature being the source of heat for reheating
-------
29
the tail gas from the primary absorber. For new acid plants, the dual
absorption system can be integrated into the plant as shown in Figure 10.*
For existing acid plants several types of add-on dual absorption systems
can be used. Figure 11* shows a flow sheet for the most generally
applicable version, this being one in which the necessary heat is obtained
by burning additional fuel. This system is independent of the existing
plant so far as heat is concerned.
For some types of acid plants it is possible to obtain at least
some of the heat for the add-on dual absorption system by integration
with the existing plant. One option is to install heat exchangers such
that the tail gas can receive heat from the interstage coolers of the
converter. Since this heat was previously used to generate steam, less
steam will now be produced and the heat cost for the dual absorption
system is reflected as a steam cost. Another option is to heat the tail
gas by mixing in some gas bypassed from the sulfur burner. This gas will
be at a temperature of 1600-1900 F and will contain 10-12 percent SO..
Because of the additional S0_, two conversion stages are required in this
case instead of one. Also, the danger of upsets is greater for this
option. Whether either of these options can be used depends on the specifics
of the acid plant. They cannot be used in wet gas acid plants because
either the heat is not available or the gas from the burner is too wet
to be introduced into the converter.
Other Emission Control Processes
A rather thorough review of all the possible emission control
systems for sulfuric acid plant tail gas has been made by Chemical
(2)
Construction Corporation. In this review, selections were made of
the processes considered "feasible" for three different levels of S02
control, i.e., effluent S02 concentrations of <100 ppm, <250 ppm, and
<500 ppm. Table 8 is an adapted listing of these selected processes which
have not been completely demonstrated but are potentially feasible based
on expected performance.
* Based on Reference 2.
-------
STEAM
SULFUR
F URNACL
SULFUH •
?PR I MARY
ABSORPTION
TOWER
JAIL
GAS
SECONDARY
ABSORPTION
' TOWER
COOLER
PRODUCT
ACID
FIGURE 10. FLOW SHEET OF DUAL ABSORPTION SYSTEM FOR NEW ACID PLANTS
-------
ABSORPTION
TOWER
EXISTING
PLANT
ADDON DUAL
ABSORPTION SYSTEM
USING EXTERNAL
HEAT SOURCE
ABSORPTION
TOWER
COMBUSTION
CHAMBER
u>
PRODUCT
ACID
FIGURE 11. FLOW SHEET OF ADD-ON DUAL ABSORPTION PROCESS USING EXTERNAL HEAT SOURCE
-------
32
TABLE 8. "FEASIBLE" EMISSION CONTROL PROCESSES FOR SULFURIC ACID PLANT TAIL GAS
(2)
Adapted from List by Chemical Construction Corp.
Processes Feasible for <100 ppm SO Control
• Adsorption on molecular sieves or resins
(a)
e Absorption by sodium sulfite-bisulfite system (Wellman-Lord)
• Absorption by other sulfite-bisulfite systems (potassium, ammonium,
methylammonium)
e Absorption by magnesium oxide
• Absorption by Na.CO to produce Na_SO
e Absorption by lime
0 Two-stage absorption by sulfuric acid and lime to recover S07 and
produce plaster of Paris
Processes Feasible for 100-250 ppm SO Control
• Absorption and oxidation of SO,, in charcoal beds (Sulfacid process)
e Absorption with basic aluminum sulfate solution (Hardman-Holden)
Processes Feasible for 250-500 ppm SO Control
e Dual absorption
(a) This control scheme was listed by Chemical Construction Corp., as
being feasible for <100 ppm SO- control. However, as discussed in
this report, data on the Wellman-Lord process indicate that the
range of effluent concentrations is 160-210 ppm S02.
-------
33
Scrubbing with an atnmoniacal solution is a control process which
has received some attention. In this process the absorption step produces
ammonium bisulfite, sulfite, and sulfate. The resulting liquid is acidified
with sulfuric acid and then air stripped to liberate the absorbed S0_ for
recycle to the acid plant. The resulting ammonium sulfate solution can
be incorporated into a diammonium phosphate fertilizer process or can be
incinerated to water vapor, nitrogen, and sulfur oxides. The major problem
with this process is that, even with mist eliminators, it is very difficult
to avoid a highly visible plume resulting from fine particles of ammonium
sulfite and bisulfite in the scrubbed tail gas.
The three processes considered in this report are generally the
most advanced processes for this application and would probably be the
most attractive at this time to the operators of sulfuric acid plants. All
three processes are currently being used, or at least tested, on commercial
acid plants.
ECONOMIC EVALUATION OF MOLECULAR SIEVE CONTROL TECHNOLOGY
In this section the economic feasibility of molecular sieve
emission control technology for sulfuric acid plants will be assessed by
comparison with the two other control processes described in the previous
section. This comparison will be based on the concept of capitalized
costs. The following will review the basis and procedures for the cost
calculations, the sources of cost data used, and the results of the
calculations.
Basis and Procedures
Investments
The investments (capital costs) for the control alternatives
were obtained from data on units which have been built and other data provided
by system designers, vendors, and operators. The costs reported are mid-1974
estimates as updated by the Marshall and Stevens Index (= 398 in mid-1974).
All calculations were made for two acid plant sizes, 200 and 1000 tons/day
of 100 percent ti^SO,. For convenience in further calculations the investments
-------
34
have been expressed by an equation of the form
•n
Investment = A(tons/day H2SOA^
where A and B are constants.
The investments include not only on-site facilities but also off-
site facilities associated with utilities. The off-site investments are
based on the following unit costs.
Electricity $360/kw
\
Cooling water $27.50/gpm
Steam $10.75/lb/hr
These values were determined in connection with a recent Battelle study.
Operating and Maintenance Costs
The operating and maintenance (0 & M) costs were calculated on
an annual basis and include the following
e Operating labor
e Utilities (electricity, cooling water, steam, and fuel)
e Raw materials and chemicals
• Maintenance labor and materials
e Local property taxes and insurance
9 By-product credits.
Table 9 lists the unit costs used in these calculations.
Again, the calculations were made for the two acid plant sizes.
For convenience in further calculations the operating and maintenance costs
have been expressed by an equation of the form
0 & M Cost = C(tons/day H^O^)3 + D(tons/day H2S04)E (2)
where B, C, D, and E are constants. The first term here is the investment-
related 0 & M costs, which includes maintenance and local taxes and insurance.
Since this is taken as a percentage of the investment, the constant B is the
same as the investment equation. The second term above includes all the
other 0 & M costs.
-------
35
TABLE 9. UNIT COSTS USED IN OPERATING
AND MAINTENANCE COSTS
Item Unit Cost
Electricity 1.2
-------
36
Capitalized Costs
The economic comparison of the control alternatives was done
through the concept of capitalized costs. The basic principle underlying
this method of economic evaluation of alternatives is the indefinite and
perpetual replacement of a given capital investment which has a finite
life. The capitalized cost is defined as the original investment plus the
present value of the perpetuity which will permit the perpetual replace-
ment of the item. For an item with an initial cost of I and a salvage
value of S after n years the capitalized cost K (in present dollars) is
— (•*• ~ s)
where i is the fractional annual interest rate or, more precisely, the rate
of return on investment which the investor requires. This is usually of
(12)
the order of 15 percent. In this study, the capitalized cost associated
with the investment, which will be designated K , was calculated from
Equation 3 with the assumption of zero salvage value. That is,
The concept of capitalized costs can be extended to include
annual and other periodic costs in addition to capital investments. An
expression for the capitalized cost associated with the annual 0 & M costs
can be obtained from Equation 3 by treating this like an investment with
a life of 1 year but discounting to present dollars, since these are end-
of-year costs. Thus,
f\ *• »*
(5)
The total capitalized cost is the sum of the term for the
initial investment and that for the 0 & M costs.
Ktot - KI + K0 4 M (6)
In this study a 15 percent rate of return and a 12-year equipment life were
used.
-------
37
Investment and Operating Costs
Molecular Sieve Process
The on-site investment and operating requirements for the PuraSiv
S process were based on a series of plots obtained from Union Carbide in
February of 1975.* These items were plotted against the tail gas flow rate.
Based on the data previously discussed, the tail gas production was taken
as 85,700 scf per ton of H.SO, for dry gas plants and 1.5 times that value
for wet gas plants. The stated accuracy of the investment costs was ±25 percent.
Other plots gave the annual cost for the adsorbent contract and the require-
ments of electrical power, heat, and cooling water. Separate plots were
given for the acid-dried air and the sieve-dried air regeneration systems.
The 0 & M costs were calculated at two different levels of control,
which were maximum effluent SCL concentrations of 100 and 500 ppm. The
level of control affects only the adsorbent contract cost and the SO,, by-
product credit. The Union Carbide plots gave the adsorbent contract cost at
both levels. The 100 ppm level corresponds to SO- emissions of about 1.5
Ib S02/ton acid for dry gas plants or 2.3 Ib S02/ton acid for wet gas plants.
These emissions are less than the New Source Performance Standard (4 Ib S09/
ton acid). The 500 ppm level corresponds to emissions greater than the New
Source Performance Standard and most state regulations but was carried
through the calculations because it provides a tie-in with the dual absorp-
tion process which cannot go below the 500 ppm level. The adsorbent contract
cost is the only cost item affected by the level of SO control.
The operating labor requirements for the molecular sieve process
are known to be very low and were taken as 1000$/yr for a 200 T/D acid
plant and 2000$/yr for a 1000 T/D acid plant.
Table 10 shows the investment and 0 & M costs for the PuraSiv
S process.
Wellman-Lord Process
The investments for the Wellman-Lord process were based on the
investment for the unit installed at the sulfuric acid plant of the Olin
Corporation at Curtis Bay, Maryland. Information on this installation was
obtained during a recent visit to the plant by Battelle personnel. '
This unit processes the tail gas from dry gas acid plant facilities with a
* These plots are presented in Appendix B. It should be noted that Union
Carbide claims that the costs of the PuraSiv S process have been further
reduced as a result of recent development work.
-------
TABLE 10. COSTS OF PURASIV S PROCESS
8000 hours/yr Operating Time
Type of H SO Plant
Regeneration Gas Drying Agent
Capacity of H SO Plant (T/D)
Tail Gas Flow Rate
(103 ft3/min at 60 F and 1 atm)
Investment (1000$, mid- 19 74)
Battery limits including license
fee
Off sites
Electricity (360$/kw)
Cooling water (27.5$/gpm)
Site preparation
TOTAL
Operating and Maintenance Cost
(1000$/yr) w
Operating labor
Molecular sieve contract
Electricity (1.2«?/kwhr)
Cooling water (30/1000 gal)
Fuel oil (12$/bbl)
Subtotal
Maintenance (2.5 percent of
inv/yr)
Taxes and insurance
(2.5 percent of inv/yr)
Total Gross
Credit for SO. recovered
(10$/T)
Total Net
Net Operating and Maintenance
Cost in $/T Total H2SO,
200
12.6
1,083
110
7
1
1,201
1.0
66.8
29.4
3.8
16.0
117.0
30.0
30.0
177.0
-13.3
163.7
2.38
Dry Gas
Acid
1,000
63.0
2,681
551
36
5
3,272
2.0
( 47.2) 334.0 (236.0)
146.9
19.0
80.2
582.1
81.8
81.8
(157.4) 745.7 (647.7)
(-11.2) -66.3 (-56.1)
(146.2) 679.4 (591.6)
(2.14) 1.98 (1.73)
200
18.9
1,565
203
13
1
1,782
1.0
118.5
54.2
6.8
46.5
227.0
44.5
44.5
316.0
-26.0
290.0
4.11
Wet Gas
Molecular Sieve
1,000
94.5
3,516
1,016
65
5
4,602
2.0
( 90.7) 592.6
270.9
33.8
232.6
1,131.9
115.0
115.0
(288.2) 1,361.9
(-22.9) -130.0
(265.3) 1,231.9
(3.78) 3.49
( 453.6)
(1,222.9)
( -114.7)
(1,108.2)
(3.16)
u>
00
(a) Main entry is for 100 ppm SO- average effluent and 2-year sieve life. Values in parentheses are for 500
ppm average effluent and 3-year sieve life.
-------
39-
total capacity of 1000 tons/day of H SO The investment for the Wellman-
Lord unit in mid-1972 was 2.8 million dollars, including the purge stream
treating facility, site preparation, control room, instrumentation,
contractor fee, etc. This investment was updated to mid-1974 using the
Marshall and Stevens Index. Based on other information available on the
(14)
Wellman-Lord Process, half the investment was assumed to depend on
the tail gas rate and half on the sulfur rate. Size exponents of 0.6 were
used for scaling both portions of the investment. In this way the investments
were calculated for the smaller plants and the wet gas plants.
The operating labor requirement was taken as one man per shift ,
based on the Olin plant. The utility and chemical requirements were based
on values given by Schneider. These values are as follows.
Value for 800 T/D
Acid Plant Pertinent Ratio
Electrical power 227 kw 0.0751 kwhr/10 scf tail gas
3
Cooling water 560 gal/min 11.1 gal/10 scf tail gas
Steam 11,000 Ib/hr 3.62 lb/103 scf tail gas
NaOH 2.65 tons/day 0.332 ton/ton S recovered
These requirements are in general agreement with other values given for
the Wellman-Lord process, ' as is shown in Appendix B.
The NaOH consumption and sulfur credit were based on 94 percent
removal of S02 from the tail gas, since this is the maximum removal achieved
by the Olin plant. Under the assumptions of this study, this removal
corresponds to the following effluent SO concentrations and emissions.
Type of acid plant Dry gas Wet gas
ppm S02 in effluent 162 210
Ib S02 emitted/ton H2SO, 2.48 4.81
Table 11 shows the investments and 0 & M costs for the Wellman-
Lord Process.
-------
40
TABLE 11. COSTS OF WELLMAN-LORD PROCESS
8000 hours/yr Operating Time
Type of H SO, Plant
^ 1
Capacity of H SO, Plant (T/D)
Tail Gas Flow Rate
(103 ft3/min at 60 F and 1 atm)
Investment (1000$, mid-1974)
Battery limits
Off sites
Electricity (360$/kw)
Steam ($10. 75/lb/hr)
Cooling water ($27.50
TOTAL
Operating and Maintenance Cost
(1000$/yr)
Operating labor, (8$/hr)
NaOH (150$/T)U;
Electricity (1.2c/kwhr)
Steam (1.2$/1000 Ib)
Cooling water (30/1000 gal)
Subtotal
Maintenance (2.5 percent of
inv/yr)
Taxes and insurance
(2.5 percent of inv/yr)
Total Gross
Credit for S recovered
J a 1
\ *y T / ^ J
Total Net
Net Operating and Maintenance
Cost in $/T H2S04
Dry
200
12.6
1,278
20
29
4
1,331
64.0
32.2
5.5
26.3
2.0
130.0
33.3
33.3
196.6
-6.5
190.1
2.85
Gas
1,000
63.0
3,357
102
147
~-^^^^^Lr*
3,625
64.0
161.0
27.3
131.5
10.1
393.9
90.6
90.6
575.1
-32.4
542.7
1.63
Wet
200
18.9
1,772
31
44
6
1,853
64.0
62.6
8.2
39.5
3.0
177.3
46.3
46.3
269.9
-12.6
257.3
3.86
Gas
1,000
94.5
4,654
153
221
29
5,057
6410
312.9
40.9
197.3
15.1
630.2
126.4
126.4
883.0
-62.9
820.1
2.4(
(a) Based on 94 percent removal of SO. from tail gas.
-------
41
Dual Absorption Process
Equations expressing the investment for the dual absorption
process as a function of the acid plant capacity were developed by Battelle
(3)
personnel as part of the Cost of Clean Air study. For new acid plants "
this investment is the difference between the investment for a dual
absorption plant and that for a single absorption plant. For existing plants
it is the investment for the add-on system using an external heat source,
although the source of heat does not greatly affect the investment. The
equation used is
Investment ($) = A(tons/day 100 percent H.SO,;.
The constant A was updated from mid-1973 to mid-1974 using the Marshall and
Stevens Index. The resulting values of the constants are as follows:
Type of Acid Plant Constant A Constant B
Dry gas, new 14,069 0.59
Dry gas, existing 47,564 0.60
Wet gas, new 30,351 0.58
Wet gas, existing 90,017 0.61
The utility requirements for the dual absorption process were
/alues
are as follows.
based on values given by Schneider for fuel-fired units. These values
Value for 800 T/D
Acid Plant
Electrical power 978 kw
Cooling water 580 gal/min
Fuel oil 99 gal/hr
These requirements were taken as directly proportional to the acid plant
capacity.
A dual absorption acid plant requires very little operating
labor beyond that for a single absorption plant. The operating labor
requirements were taken as 1000$/yr for a 200 T/D plant and 2000$/yr for
a 1000 T/D plant, the same values as were used for the molecular sieve process.
-------
42
Table 12 shews the investments and 0 & M costs for the dual
absorption process for new acid plants and Table 13 the corresponding costs
for existing acid plants.
Summary of Costs and Equations
Table 14 summarizes the investments and operating costs for the
three emission control processes. Values are given for the molecular sieve
process at two levels of control (100 and 500 ppm), for two acid plant
sizes (200 and 1000 tons/day), and for the two types of acid plants (dry
gas and wet gas). In addition to the investments and 0 & M costs taken
from the previous tables, Table 14 includes the total operating cost
obtained by adding to the 0 & M cost the capital-related costs for depreciation
and return on investment.
Table 15 presents the equations for the total investment and net
0 & M costs as functions of acid plant capacity for the molecular sieve,
Wellman-Lord, and dual absorption processes.
Capitalized Costs
Table 16 shows the capitalized costs for the three emission
control processes considered here. As in the previous table, values are
given for the molecular sieve process at two levels of control, for two
acid plant sizes, and for the two types of acid plants.
Figures 12 and 13 show the capitalized costs for dry gas and
wet gas acid plants, respectively, as functions of capacity. These figures
were prepared from the values in Table 16 plus values calculated for two
other plant sizes (400 and 2000 tons/day).
For dry gas acid plants, the dual absorption process is the
least expensivei However, it should be noted at this point that this
process is also the least effective, it being limited to an effluent con-
centration of about 500 ppm S0«. This concentration corresponds to
emissions greater than the New Source Performance Standard and most state
regulations. Of the other two processes, the molecular sieve process
is less expensive for smaller acid plants (less than about 700 tons/day)
and the Wellman-Lord process is less expensive for larger acid plants.
-------
43
TABLE 12. COSTS OF DUAL ABSORPTION PROCESS
FOR NEW ACID PLANTS
8000 hours/yr Operating Time
Type of H0SO. Plant
Z H -
Capacity of H SO, Plant (T/D)
Tail Gas Flow Rate
(10 ft /min at 60 F and 1 atm)
Investment (1000$, mid- 19 74)
Battery limits
Off sites
Electricity (360$/kw)
Cooling water (27.5$/gpm)
TOTAL
Operating and Maintenance Cost
(1000$/yr)
Operating labor
Electricity (1.2c/kwhr)
Cooling water (3C/1000 gal)
Fuel oil (12$/bbl)
Subtotal
Maintenance (4.0 percent of
inv/yr)
Taxes and insurance
(2.5 percent of inv/yr)
Total Gross
Credit for additional H.SO,
(25$/T)w *
Total Net
Net Operating and Maintenance
Cost in $/T Total H0SO.
2 4
Dry
200
12.6
321
88
4
413
1.0
23.4
2.1
56.7
83.2
16.5
10.3
110.0
-42.9
67.1
0.98
Gas
1,000
63.0
828
440
20
1,288
2.0
117.2
10.4
283.3
412.9
51.5
32.2
496.6
-214.6
282.0
0.82
Wet
200
18.9
656
132
6
794
1.0
35.2
3.2
85.0
124.4
31.8
19.9
176.1
-87.8
88.3
1.26
Gas
1,000
94.5
1,668
659
30
2,357
2.0
175.8
15.6
425.0
618.4
94.3
58.9
771.6
-439.0
332.6
0.95
(a) Based on 500 ppm SO,, in treated tail gas.
-------
44
TABLE 13. COSTS OF ADD-ON DUAL ABSORPTION PROCESS
FOR EXISTING ACID PLANTS
8000 hours/yr Operating Time
Type of H SO, Plant
Capacity of H-SO Plant (T/D)
Tail Gas Flow Rate
(10 3 ft3/min at 60 F and 1 atm)
Investment (1000$, mid- 19 74)
Battery limits
Off sites
Electricity (360$/kw)
Cooling water (27.5$/gpm)
TOTAL
Operating and Maintenance Cost
(1000$/yr)
Operating labor
Electricity (1.2c/kwhr)
Cooling water (3
-------
TABLE 14. INVESTMENT AND OPERATING COSTS FOR EXISTING ACID PLANTS
Control Process
ppm SO in Effluent
Size and Type of Plant
200 T/D dry gas
1000 T/D dry gas
200 T/D wet gas
1000 T/D wet gas
Cost Iten/a)
Investment (M$)
Operating cost (M$/yr)
0 & M f .
Capital (b)
TOTAL
Investment (M$)
Operating cost (M$/yr)
0 & K CM
Capital (b)
TOTAL
Investment (M$)
Operating cost (M$/yr)
0 & M rM
Capital00
TOTAL
Investment (M$)
Operating cost (M$/yr)
0 & M IM
Capital0*'
TOTAL
Molecular Sieve
100
1,201
163.7
221.6
385.3
3,272
679.4
603.6
1,283.0
1,782
290.0
328.7
618.7
4,602
1,231.9
849.0
2,080.9
Molecular Sieve
500
1,201
146.2
221.6
367.8
3,272
591.6
603.6
1,195.2
1,782
265.3
J28.7
594.0
4,602
1,108.2
849.0
1,957.2
Wellman-Lord
160-210
1,331
190.1
245.5
435.6
3,625
542.7
668.7
1,211.4
1,853
257.3
341.8
599.1
5,057
820.1
932.9
1,753.0
Dual Absorption
500
1,235
120.6
227.8
348.4
3,461
423.2
638.5
1,061.7
2,418
193.8
446 ._!
639.9
6,775
619.8
1,249.9
1,869.7
(a) M = 1000.
(b) Includes return on investment at 15 percent/yr and sinking-fund depreciation for 12-year life at 15 percent/yr
interest.
Ln
-------
TABLE 15. EQUATIONS FOR INVESTMENT AND OPERATING COSTS
Control Process
Molecular sieve
Molecular sieve
Molecular sieve
Molecular sieve
Wellman-Lord
Wellman-Lord
Dual Absorption
Dual Absorption
Type of
Acid Plant
Dry gas
Wet gas
Dry gas
Wet gas
Dry gas
Wet gas
Dry gas
Wet gas
Total Investment
(1000$) (a) Net Operating + Maintenance Cost ($/yr)
44.26 C°'623
78.66 C°'589
44.26 C°'623
78.66 C°'589
49.03 C°'623
67.91 C°'624
41.60 C°-64°
81.44 C°-64°
526. 7 C°-"7
1,015.8 C '
440.1 C '
890. 7 C°-"8
3,606 C°'667 +
2,816C°-768+
212.5 C°-99°
195.0 C°'988
+ 2,213 CC'623
+ 3,933 C°'589
+ 2,213 C°'623
+ 3,933 C0'589
2,452 C°'623
3,396 C°'624
+ 2,704 C°-64°
+ 5,294 C°-64°
ppm S02 in
Effluent Gas
100
100
500
500
162
210
500 .
500
(a) Includes offsites.
(b) Add-on process for existing acid plant.
-------
TABLE 16. CAPITALIZED COSTS FOR EXISTING ACID PLANTS
Capitalized Cost (1000$)
Control Process
ppin S00 in Effluent
A.
Size and Type of Plant
200 T/D dry gas
1000 T/D dry gas
200 T/D wet gas
1000 T/D wet gas
Molecular sieve
Invest.
O&M
TOTAL
Invest.
O&M
TOTAL
Invest.
O&M
TOTAL
Invest.
O&M
TOTAL
100
1,477
1.091
2,568
4,025
A. 5 29
8,554
2,192
1.933
4,125
5,660
8.213
13,873
Molecular sieve
500
1,477
975
2,452
4,025
3.944
7,969
2,192
1,769
3,961
5,660
7.388
13,048
Wellman-Lo rd
160-210
1,637
1.267
2,904
4,458
1.618
8,076
2,279
1.1.715
3,994
6,219
5,467
11,686
Dual Absorption
500
1,519
804
2,323
4,257
2,821
7,078
2,974
1.292
4,266
8,332
4,132
12,464
Note: Investment capitalized cost based on 12-year life; both investment and O&M capitalized costs
based on 15 percent/year "interest" rate.
-------
48
100 9.
•w-
vO
o
0}
o
u
•a
a)
N
0
2.5 3
5 6
6000
H.SO Plant Capacity (tons/day)
FIGURE 12. CAPITALIZED COSTS FOR SO- CONTROL OF DRY GAS ACID PLANTS
-------
49
100
6000
H SO, Plant Capacity (tons/day)
FIGURE 13. CAPITALIZED COSTS FOR SO CONTROL OF WET GAS ACID PLANTS
-------
50
Again, it should be noted that the molecular sieve process is capable of
a higher degree of SCL removal than is the Wellman-Lord process (<100
ppm versus 160-210 ppm S0?), although both can meet the New Source Perfor-
mance Standard for most plants.
For wet gas acid plants, the Wellman-Lord process is the least
expensive. Dual absorption has the next lowest cost except for very small
acid plants, where the molecular sieve process becomes slightly less
expensive.
The accuracy of the costs used here does not permit precise
definition of the economic breakpoints, particularly since the costs for
the three processes are quite close at all capacities. The point is that
the costs are close enough that the molecular sieve process appears to be
economically competitive, at least for the smaller acid plants.
-------
51
CONCLUSIONS AND RECOMMENDATIONS
This analysis of the applicability of molecular sieve technology
to the control of SO. emissions from sulfuric acid plants leads to the
following conclusions: . .
e Molecular sieve technology can control the S0_
emissions to well within the most stringent
existing regulation for sulfuric acid plants (4
Ib SO /ton acid). Union Carbide field test data
indicated that shortly after startup the Coulton
unit was limiting the emissions to no more than
0.4 Ib SO./ton acid. The data taken by York
Research Corporation after the equivalent of 10
months of operation on the second charge of sieve
showed an average emission of 0.9 Ib SO./ton
acid. Union Carbide's performance guarantee for
this unit corresponds to an emission of about 1.2
Ib SO./ton acid.
• The PuraSiv S process can limit the S0? emissions
more effectively than either of the otfier processes
being used or tested commercially on acid plants.
The average effluent SO- concentration is <100
ppm, compared to 160-210 ppm for the Wellman-Lord
process and about 500 ppm for the dual absorption
process.
0 The S09 emissions from a molecular sieve control
unit increase with increasing inlet (tail gas)
S02 concentration. At the Coulton unit the average
emission for the cycles in which the inlet S0?
concentration exceeded 4500 ppm was 1.2 Ib S0?/
ton acid.
e The assumed 2-year sieve life with acceptable S0_
control has not been demonstrated, although there
is no reason to believe it cannot be achieved. The
field tests to date have been too short to detect
any change in the emissions with time.
• Molecular sieve technology appears to be economically
competitive with the Wellman-Lord and dual absorption
processes, at least for the smaller acid plants. The
capitalized costs for all three processes are fairly
close, but the dual absorption process is the least
expensive for dry gas acid plants and the Wellman-Lord
process is the least expensive for wet gas plants.
However, the dual absorption process is limited in
its effectiveness such that for most plants it cannot
meet the regulation of 4 Ib SO /ton acid (although it
-------
52
can meet some less stringent regulations). Individual
acid plant characteristics will affect the economic
choice among these alternates.
The only recommendation arising from this study is that the 2-year
sieve life be demonstrated by monitoring the SO- emissions at a commercial
unit periodically over that time period.
-------
53
REFERENCES
(1) U.S. Bureau of Census, Current Industrial Reports Series M28A (71) -
13 (1970-1974).
(2) Chemical Construction Corporation, "Engineering Analysis of Emissions
Control Technology for Sulfuric Acid Manufacturing Processes", Report
No. PB 190393, March, 1970.
(3) "Cost of Clean Air, 1975", report in progress by Battelle-Columbus.
(4) "Atmospheric Emissions from Sulfuric Acid Manufacturing Processes",
Public Health Service Publ. 999-AP-13, 1965.
(5) Federal Register, J36, 24881, December 23, 1971.
(6) Breck, D. W., Zeolite Molecular Sieves. Wiley, 1974.
(7) Martin, D. A. and Brantley, F. E., "Selective Adsorption and Recovery
of Sulfur Dioxide from Industrial Gases by Using Synthetic Zeolites",
Bureau of Mines Report of Investigations 6321, April, 1963.
(8) Collins, J. J., et al., "The PuraSiv S Process for Removing SO from
Sulfuric Acid Plant Tail Gas", AIChE meeting, Philadelphia, Pennsylvania,
November 15, 1973.
(9) York Research Corporation, "Evaluation of H-SO, Plant Emissions at
the Coulton Chemical Corporation Which is Controlled by the PuraSiv
S Process", EPA Contract No. 68-02-1401, Task 2, Report No. Y-8479-2,
May 6, 1975.
(10) Study in progress for American Petroleum Institute.
(11) Peters, M. S. and Timmerhaus, K. D., Plant Design and Economics for
Chemical Engineers. 2nd Edition, McGraw-Hill, pp 173-175, 1968.
(12) Clapperton, J. A., PPG Industries, Pittsburgh, Pennsylvania, letter
to M. Y. Anastas of BCL, February 28, 1975.
(13) Murthy, K. S. and Rosenberg, H. S., Draft Report on April 29, 1975
trip to Olin Corporation, Curtis Bay, Maryland.
(14) McGlamery, G. G. and Torstrick, R. L., "Cost Comparisons of Flue
Gas Desulfurization Systems", Symposium on Flue Gas Desulfurization,
Atlanta, Georgia, November 4-7, 1974 (and EPA 650/2-74-126-a,
December, 1974).
(15) Schneider, R. T., "Three Solutions to the SO. Emission Problem from
Sulfuric Acid Plants", AIChE meeting, Daytona Beach, Florida,
May 19, 1973.
(16) Cassetty, R. W., Union Carbide Corporation, Tarrytown, New York,
personal communication to D. W. Hissong of BCL, August 1, 1975.
-------
APPENDIX A
ACCURACY OF SO,. MEASUREMENTS
-------
APPENDIX A
ACCURACY OF SO MEASUREMENTS
This Appendix presents and expands upon an analysis by York
Research Corporation of the accuracy of their S09 concentration measurements.
These measurements were made using a Du Pont 460/1 photometric analyzer. The
accuracy of this analyzer was assessed by comparing its results with the
results of simultaneous measurements using EPA Method 6.
Table A-l presents the data from the accuracy tests. Included
are seven points for inlet samples with S0? concentrations of 2500-3400 ppm
and seven points for outlet samples with S0? concentrations of 10-135 ppm.
For each point, the absolute difference between the two analytical methods,
D, and its square are tabulated.
Statistical calculations were made for three data sets - the seven
inlet samples, the seven outlet samples, and all 14 samples together. For
each set, the average absolute deviation was calculated as
n
1 E
Average Absolute Deviation = D = — . = . D.
where n is the number of data points. The root mean square deviation was
calculated as
jT n 2 11/2
Root Mean Square Deviation » — I D.
The 95 percent confidence interval was calculated as
D
1 J
where t is the 2.5 percent point from the "t distribution" for n - 1 degrees
of freedom. For n = 7 the value of t is 2.447 and for n = 14 it is
2.160. The accuracy of the Du Pont measurements, expressed as a percentage,
was then calculated as
-------
A-2
TABLE A-l. SO. MEASUREMENT ACCURACY TESTS BY YORK RESEARCH
vppm SO,, by
Location Date
Inlet 2/12
>
2/12
2/14
2/14
2/17
2/17
' 2/18
Outlet 2/12
\
2/14
2/14
2/17
2/17
2/17
' 2/18
i
Time
1035
1555
0924
0925
1133
1320
1110
1147
1134
1224
1135
1340
1435
1.116
Test No.
1
3
4A
4B
7
9
13
2
5
6
8
10
11
12
EPA Method 6 "" Du Pont
3322
3359
3176
3022
2615
2749
3354
72.3
115.
15.1
88.3
16.2
45.7
120.6
3300
3385
3217
3217
2500
2525
3400
Sum
73.
102.
• 7.5
84.
10.5
39.
135.
Sum
Grand Sum
Difference _
D D^
22
26
41
195
115
224
46
669
0.7
13.
7.6
4.3
5.7
6.7
14.4
52.4
721.4
484
676
1,681
38,025
13,225
50,176
2,116
106,383
0.49
169.
57.76
18.49
32.49
44.89
207.36
530.48
106,913.48
-------
A-3
(D + C )(100)
Accuracy (%) - — —
1 n
Z V.
n i = 1
where V is the value (SO concentration) by EPA Method 6. This definition
is the one specified in the Federal Register* for SO. emissions testing.
The results of the statistical calculations are shown in Table
A-2. The accuracy of the Du Pont analyzer is about 6 percent for the high
concentration data and for the data as a whole, but about 18 percent for the
low concentration data alone. Even this value is less than the 20 percent
accuracy limit specified in the Federal Register.
* "Performance Specifications and Specification Test Procedures for Monitors
of SO and NO from Stationary Sources", Federal Register, J9, (177),
September 11,X1974.
-------
A-4
TABLE A-2. S02 MEASUREMENT ACCURACY PARAMETERS - YORK RESEARCH TESTS
Type of Samples
S0» Concentration Range (ppm)
Number of Samples Compared
Comparison of Du Pont Analyzer with EPA Method 6
Average absolute difference (ppm)
Root mean square difference (ppm)
95 percent confidence interval (ppm)
Accuracy (percent)
Inlet
2500-3400
7
95.6
46.6
77.8
5.62
Outlet
10-135
7
7.5
3.3
4.4
17.64
Inlet & Outlet
10-3400
14
51.5
23.4
42.3
5.95
-------
APPENDIX B
DETAILED COST DATA
-------
APPENDIX B
DETAILED COST DATA
This Appendix presents some details related to the cost data
discussed in the text. The first section deals with the PuraSiv S
process and the second section v,ith the Wellman-Lord process.
PuraSiv S Process
This section includes the series of plots obtained from Union
Carbide on the costs and cost-related quantities for the PuraSiv S
process. The following figures are included:
Figure Title
B-l , Investment for PuraSiv S Process with Acid-Dried
Regeneration
B-2 Adsorbent Contract Cost, Acid-Dried Regeneration, Two
Years, 100 ppm S0_
B-3 Adsorbent Contract Cost, Sieve-Dried Regeneration, Two
Years, 100 ppm S02
B-4 Adsorbent Contract Cost, Acid-Dried Regeneration, Three
Years, 500 ppm S0_
B-5 Adsorbent Contract Cost, Sieve-Dried Regeneration, Three
Years, 500 ppm S0_
B-6 Power Requirements, Acid-Dried Regeneration
B-7 Power Requirements, Sieve-Dried Regeneration
B-8 Absorbed Heat Requirements, Acid-Dried Regeneration
B-9 .Absorbed Heat Requirements, Sieve-Dried Regeneration
B-10 Cooling Water Requirements, Acid-Dried Regeneration
B-ll Cooling Water Requirements, Sieve-Dried Regeneration.
These figures present the cost-related quantities used in the
cost estimates presented in this report. However, it should be noted
that in recent communication* Union Carbide claimed that their recent
development work on the PuraSiv S process has resulted in some further
* Cassetty, R. W., Union Carbide Corporation, Tarrytown, New York,
personal communication to D. W. Hissong of BCL, August 1, 1975.
-------
B-2
3,000
(0
M
id
•g 2,000
.
-o
•o
| -1,500
0)
I
60
a
•H
O
a
I
4J
0)
I
M
0)
4J
•H
CO
1,000
800
600
500
400
300
200
I
I
Spring, 1974 Cost
M & S Index - 386
Accuracy ±25%
I
6 8 10 2Q 30 40
Tall Gas Flow Rate (thousand scfm)
60
80 100
FIGURE B-l. INVESTMENT FOR PURASIV S PROCESS WITH ACID-DRIED REGENERATION
Notes:
(1) Costs shown do not include off sites, motor control center,, site preparation
or molecular sieve adsorbent.
(2) Costs shown are for tail gas streams containing less than 3000 ppmv S0_.
For streams containing 4000 ppmv S02, add 10 percent. Not applicable to
tail gas streams containing more than 4000 ppmv SO..
(3) For a molecular sieve dried air regeneration system or for a UCC supplied
acid dried air regeneration system, add 10% to above shown costs for tail
gas streams containing less than 3000 ppmv SO,. Add an additional 10% for
streams containing 4000 ppmv SO.. Not applicable to tail gas streams
containing more than 4000 ppir.v SO-.
(4) In these figures, scfm = cubic feet per minute at 60 F and atmospheric
pressure.
8/19/74
-------
B-3
^
s
0)
a.
CO
s
rH
O
•d
•O
us an
o
•g
4J
m
0
u
4J
u
0)
a
8
4J
C
-------
B-4
420
400
380
360
340
g, 320
to
5 300
o
•0
-
§
•
280
260
240
o
0 220
o
cd
M
4J
(3
O
O
200
180
g
•S 160
o
140
120
100
80
60
40
20
S02 Adsorber
Section
10
20 30 40 50 60
Tail Gas Flow Rate (thousand scfm)
70
80
FIGURE B-3. ADSORBENT CONTRACT COST, SIEVE-DRIED REGENERATION,
TWO YEARS, 100 PPM SO-
12/12/74
-------
B-5
at
o.
380
360
340
320
300
280
260
240
o
•a 220
•a ;
I 200
o
£ . 180
160
2 ' 140
§
0 , 120
4J
01
•e -"ioo
o
3
< .80
60
40
20
0
10
I
1
I
I
20 ,; 30,, 40 50 60
Tail Gas Flow Rate (thousand scfm)
70
80
FIGURE B-4. ADSORBENT CONTRACT COST, ACID-DRIED REGENERATION,
THREE YEARS, 500 PPM SO,
12/12/74
-------
B-6
340 (
320
300
3 280
>> •
u
o.
CO
'3
260
240
•8 220
200
180
CO
o
u
4-1
o
a
a
o
o
160
140
120
S
•B 100
o
•s
< 80
60
40
20
10 20 30 40 50
Tail Gas Flow Rate (thousand scfm)
60
70
80
FIGURE B-5. ADSORBENT CONTRACT COST, SIEVE-DRIED REGENERATION,
THREE YEARS, 500 PPM SO.
12/12/74
-------
B-7
2
2000
1800
]600
1400
1200
1000
M
41
g 800
600
400
200
Accuracy ±25%
I
I
I
I
10 20 30 40 50
Tail Gas Flow Rate (thousand scfm)
60
70
80
FIGURE B-6. POWER REQUIREMENTS, ACID-DRIED REGENERATION
10/5/73
-------
B-8
a
>j
2800
2600
2400
2200
2000
1800
1600
g. 1400
a>
1200
1000
800
600
400
200
0
Accuracy ±25%
I
I
I
I
_L
10
20 30 40 50
Tail Gas Flow Rate (thousand scfm)
60
70
80
FIGURE B-7. POWER REQUIREMENTS, SIEVE-DRIED REGENERATION
10/5/73
-------
B-9
5.0
4,5
4.0
*• 3>5
4J
W
°S . 3.0
v^
CO
U
C
2 2.5
•o
01
•e
o
CD
on
2.0
1.5
1.0
0.5
I
Accuracy ±15 percent
I
I
I
10 ^20 30 40 50
Tail Gas Flow Rate (thousand scfm)
60
70
80
FIGURE B-8. ABSORBED HEAT REQUIREMENTS, ACID-DRIED REGENERATION
-------
B-10
10.0
9.0
8.0
7.0
6.0
!• 5.0
-------
B-ll
1800
1600
1400
H
<
1200
a
1000
800
-I
oo
0
o
5
600
400
200
I
I
I
I
I
Accuracy ±15%
I
I
„ 10
60
70
., 20, .. .30 40 50
Tall Gas Flow Rate (thousand scfm)
*
FIGURE B-10. COOLING WATER REQUIREMENTS, ACID-DRIED REGENERATION
80
10/5/73
-------
B-12
H
<
0
SS
N—/
s
2000
1800
1600
1400
1200
4)
| 1000
I 800
DO
a
o
o
o
600
400
200
I
I
I
10
20 30 40 50
Tail Gas Flow Rate (thousand scfm)
60
70
80
FIGURE B-ll. COOLING WATER REQUIREMENTS, SIEVE-DRIED REGENERATION
-------
B-13
decreases in the process cost. These additional improvements could not
be factored into this study but should be kept in mind when comparisons
with other technology options are made.
i ... ' •'
The Coulton PuraSiv S unit is not a good basis for analyzing
the investment required for this process since it is not a typical
installation. However, it can be used for an approximate check of the
investment level shown in Figure B-l. When the Coulton unit was built
(8)
in 1972, the investment was about.$397,000. This cost was low because
(1) It did not include design, engineering, or con-
tractor's fees since Coulton did their own
design and construction work
(2) It did not include the tail gas blower, since
, Coulton supplied this
(3) Coulton bought used mist eliminators
(4) Union Carbide gave Coulton a low price because
this was the first-of-a-kind installation.
The following analysis leads to an approximate 1974 cost for a
complete plant. Updating the above cost from 1972 to 1974 using the
Marshall and Stevens Index gives a cost of $476,000. A blower designed for
the maximum tail gas flow rate of 10,446 scfm, a system pressure drop of
3 psia,* an inlet temperature of 70 F, and an efficiency of 60 percent
will require 232 brake horsepower. Based on the cost analysis method of
Guthrie,** the 1974 "bare module" cost of this blower would be about
$222,000. Adding this to the $476,000 and adding 15 percent for design,
engineering,' and contractor's fees gives a total investment of about
$803,000. This compares reasonably well with the cost of about $900,000
obtained from Figure B-l, the difference being attributable to Items 3 and
4 above.
A rough verification of the electrical power requirements
(Figures B-6 and B-7) can also be made from the blower design mentioned
above. The 232 brake horsepower for the tail gas blower corresponds to a
power requirement of 173 kW. Adding to this the requirements for the
regeneration air blowers, using the same 3 psi pressure drop and the air/
tail gas flow ratios from Figures 4 and 5 of the text, gives total blower
* Information obtained by Battelle personnel during visit to Coulton plant
on June 12, 1974.
** Guthrie, K. M., "Capital Cost Estimating", Chemical Engineering, p 114,
March 24, 1969.
-------
B-14
power requirements of 216 kW for an acid-dried air system and 259 kW for
a sieve-dried air system. These values are below but fairly close to the
requirements of 260 kW (acid dried) and 300 kW (sieve dried) obtained from
Figures B-6 and B-7, reflecting the use of some power for items other than
blowers.
Wellman-Lord Process
The investment for the Wellman-Lord process was based on the
unit installed at Olin Corporation's plant at Curtis Bay, Maryland. The
utility and chemical requirements were based on values given by R. T.
Schneider of Davy Powergas, Inc.,* since this was the only complete set
of requirements available. Davy Powergas is the vendor of the Wellman-
Lord process in the United States.
Since data are available from several sources, it may be help-
ful to compare these data. Table B-l compares the data from the three
sources analyzed as part of this study.
Schneider, R. T., "Three Solutions to the SO. Emission Problem from
Sulfuric Acid Plants", AIChE meeting, Daytona Beach, Florida,
May 19, 1973.
-------
B-15
TABLE B-l. COMPARISON OF COST DATA FOR WELLMAN-LORD PROCESS
Data Source
Reference Number
Type of Plant
Plant Size
3
Inlet Gas Rate (10 cfm at 60 F)
Sulfur Recovery Rate (T/D)
Operating Time (hours /yr)
Directly Derived Data
3
Investment (10 $)
Date for Investment
Electricity (kW)
Cooling Water (gal/min)
Steam (lb'/hr)
NaOH (T/D)
Other Calculated Quantities
3 (d}
1974 Investment (10 $)V
3
Electricity (kwhr/10 scf gas)
3
Cooling Water (gal/10 scf gas)
3
Steam (lb/10 scf gas)
NaOH (T/T sulfur recovered)
Schneider
15
H2S°4
800 T/D H SO
( AT
50.5
7.98(a)
7992
1,250
May, 1973
227
560
11,000
2.65
1,452
0.0751
11.1
3.62
0.332
Olin Curtis
Bay Plant
13
H2S°4
1000 T/D H SO
/ 4
67.6
16.3(b)
2,800
Mid- 19 72
342
25,868
5.75
3,357
0.0844
6.38
0.353
McGlamery &
Torstrick
14
Power plant
500 MW
1248
112.1
7000
19,184
Mid- 19 74
305,400(c)
19,184
(cY
4.08U;
(a) Based on average data for U.S.
industry, as discussed in text.
(b) Based on quoted average inlet and outlet SO^ concentrations of 4000 and 293 ppm,
respectively.
(c) Includes net steam requirement for SO reduction facility, which could be
either positive or negative.
(d) Chemical Engineering Plant Cost Index = 165.4, Marshall & Stevens Index = 398.
-------
TECHNICAL REPORT DATA
(Please read liiumctions on the reverse before completing)
1. REPORT NO.
EPA-600/2-75-066
2.
3. RECIPIENT'S ACCESSION'NO.
4. TITLE AND SUBTITLE
Molecular Sieve Control Process in Sulfuric Acid
Plants
5. REPORT DATE
October 1975
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO
D.W. Hissong
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Battelle-Columbus Laboratories
505 King Avenue
Columbus, Ohio 43201
10. PROGRAM ELEMENT NO.
1AB013; ROAP 21ADH-005
11. CONTRACT/GRANT NO.
68-02-1323, Task 17
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Task Final; 10/74-10/75
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
report gives results of an engineering analysis of the applicability of
molecular sieve technology to the control of SO2 emissions from sulfuric acid plants.
After the equivalent of 10 months of operation, one plant using this technology is still
controlling SO2 emissions to well within Federal and State regulations for sulfuric
acid plants. It is also meeting the performance guarantee of the process developer
and vendor. Although the concept of a 2-year sieve life with acceptable SO2 control
has not been demonstrated, there is no reason to believe that it cannot be achieved:
at this point this application of molecular sieve technology appears technically feas-
ible. The economic feasibility of the technology for this application was assessed by
comparing its total capitalized cost (including investment and operating cost) with
that of the Wellman-Lord and dual absorption processes. Capitalized costs for the
three are fairly close; individual plant characteristics will affect the economic
choice. The technology is more competitive for smaller plants and for those which
already have sieve -regeneration air driers. Although the dual absorption process
will be the least expensive for many plants, it is limited in its effectiveness.
Considering overall cost and effectiveness, molecular sieve technology appears to
be economically feasible for some acid plants.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COS AT I Field/Group
Air Pollution
Absorbers (Materials)
Sulfur Dioxide
Emission
Sulfuric Acid
Industrial Plants
Analyzing
Air Pollution Control
Stationary Sources
Molecular Sieves
13B
11G
07B
14B
13. DISTRIBUTION STATEMENT
19. SECURITY CLASS (ThisReport)
Unclassified
Unlimited
21. NO. OF PAGES
82
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
------- |